IP Library Granted Patent US 9,767,357
Granted Patent B2
US 9,767,357 · App. 14/983,385 · Granted Sep 19, 2017

Systems and methods for image recognition normalization and calibration

Inventors: Nikhil Johri (Mountain View, CA); Balamanohar Paluri (Menlo Park, CA); Lubomir Bourdev (Mountain View, CA)
Assignee: Facebook, Inc
G06K9/00536G06K9/00288G06K9/623G06K9/6253G06K9/6293
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,767,357
App. No.
14/983,385
Granted
Sep 19, 2017
Kind
B2
Abstract

Systems, methods, and non-transitory computer-readable media can calculate raw scores for a plurality of media items based on a classifier model and a target concept. The plurality of media items are ranked based on the raw scores. A review set of the plurality of media items is determined, the review set comprising a subset of the plurality of media items. Each of the media items of the review set is associated with a content depiction determination. A normalized score formula is calculated based on the raw scores and the content depiction determinations for the media items of the review set.

Claims (37)

1. A computer-implemented method comprising:

calculating, by a computing system, raw scores for a plurality of media items based on a classifier model and a target concept;

ranking, by the computing system, the plurality of media items based on the raw scores;

determining, by the computing system, a review set of the plurality of media items, the review set comprising a subset of the plurality of media items;

associating, by the computing system, each of the media items of the review set with a content depiction determination, wherein the content depiction determination is indicative of whether a media item depicts the target concept; and

calculating, by the computing system, a normalized score formula based on the raw scores and the content depiction determinations for the media items of the review set.

2. The computer-implemented method of claim 1 , wherein the review set is determined based on a sampling rate.

3. The computer-implemented method of claim 2 , further comprising receiving a sampling rate selection from a user.

4. The computer-implemented method of claim 1 , wherein calculating the normalized score formula comprises calculating a logistic regression formula based on the raw scores and the content depiction determinations for the media items of the review set.

5. The computer-implemented method of claim 1 , further comprising presenting a user interface configured to receive content depiction determinations for the media items of the review set.

6. The computer-implemented method of claim 1 , further comprising re-training the classifier model based on the normalized score formula.

7. The computer-implemented method of claim 6 , further comprising repeating the computer-implemented method with the re-trained classifier model.

8. The computer-implemented method of claim 1 , wherein the normalized score formula is configured to convert a raw score calculated by the classifier model into a normalized score.

9. The computer-implemented method of claim 8 , wherein the normalized score is a probability value.

10. The computer-implemented method of claim 1 , wherein the review set comprises a fixed number of media items.

11. A system comprising:

at least one processor; and

a memory storing instructions that, when executed by the at least one processor, cause the system to perform a method comprising:

calculating raw scores for a plurality of media items based on a classifier model and a target concept;

ranking the plurality of media items based on the raw scores;

determining a review set of the plurality of media items, the review set comprising a subset of the plurality of media items;

associating each of the media items of the review set with a content depiction determination, wherein the content depiction determination is indicative of whether a media item depicts the target concept; and

calculating a normalized score formula based on the raw scores and the content depiction determinations for the media items of the review set.

12. The system of claim 11 , wherein the review set is determined based on a sampling rate.

13. The system of claim 12 , wherein the method further comprises receiving a sampling rate selection from a user.

14. The system of claim 11 , wherein calculating the normalized score formula comprises calculating a logistic regression formula based on the raw scores and the content depiction determinations for the media items of the review set.

15. The system of claim 11 , wherein the method further comprises presenting a user interface configured to receive content depiction determinations for the media items of the review set.

16. A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computing system, cause the computing system to perform a method comprising:

calculating raw scores for a plurality of media items based on a classifier model and a target concept;

ranking the plurality of media items based on the raw scores;

determining a review set of the plurality of media items, the review set comprising a subset of the plurality of media items;

associating each of the media items of the review set with a content depiction determination, wherein the content depiction determination is indicative of whether a media item depicts the target concept; and

calculating a normalized score formula based on the raw scores and the content depiction determinations for the media items of the review set.

17. The non-transitory computer-readable storage medium of claim 16 , wherein the review set is determined based on a sampling rate.

18. The non-transitory computer-readable storage medium of claim 17 , wherein the method further comprises receiving a sampling rate selection from a user.

19. The non-transitory computer-readable storage medium of claim 16 , wherein calculating the normalized score formula comprises calculating a logistic regression formula based on the raw scores and the content depiction determinations for the media items of the review set.

20. The non-transitory computer-readable storage medium of claim 16 , wherein the method further comprises presenting a user interface configured to receive content depiction determinations for the media items of the review set.

Assignments (2)
CHANGE OF NAME Recorded Nov 24, 2021
From: FACEBOOK, INC.
To: META PLATFORMS, INC.
Reel/Frame 058250/0214 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 11, 2017
From: JOHRI, NIKHIL; PALURI, BALAMANOHAR; BOURDEV, LUBOMIR
To: FACEBOOK, INC.
Reel/Frame 042979/0818 →
Continuity (1)
Related Publication 20170185838A1 · Jun 29, 2017